Machine learning in trading: theory, models, practice and algo-trading - page 2685

[mytarmailS]

Maxim Dmitrievsky #:
and in such ancient ways.

In what ways?

[[Deleted]]

mytarmailS #:
By what means?

By old books such as technical analysis, martingales of all kinds, on indicators.

If you hit a trend, you make money, or vice versa.

[mytarmailS]

Maxim Dmitrievsky #:
On old books, for example, technical analysis, martingales of all sorts, on indicators.

If you hit a trend, you make money, or vice versa.

The new ones are no better.

[Roman]

Renat Fatkhullin #:

Posted documentation on the new features of the built-in matrices and vectors:

• Matrixand Vector Methods
  
• Matrix and vector types
• Initialisation
• Manipulations
• Operations
• Derivations
• Transformations
• Statistics
• Characteristics
• Solutions
• Machine learning

Please go through the descriptions of functionality.

If you can give us your comments and ideas, it would be very much appreciated.

Renat, could you please tell me if multidimensional matrices are not available?
Where the matrix represents an n-dimensional space.

A multidimensional representation would be nice.

matrix   matrix_a(4, 8, 3);

[Renat Fatkhullin]

Roman #:

Renat, can you please tell me if multidimensional matrices are not available?
Where the matrix is an n-dimensional space.

A multidimensional representation would benice.

You can use the usual standard arrays up to the 4th dimension or your own classes. But the maths would already have to be done yourself.

We've done a good job with the basic maths of vectors and matrices. If developers use them, it will cover 99% of the needs.

Of course, we will expand the set of maths further.

[[Deleted]]

mytarmailS #:
The new ones are no better.

read what programme you need to pass for quantum, find it. Write your own tester of course 😁, regression and correlation analysis, risk assessment, bonus machine learning.

Not a word about DSP again. Better or no better... this is the de facto basis now, to at least have an outlook, then the junior may be hired, to fill in reports.

[mytarmailS]

How do I know if this signal has alpha

[[Deleted]]

Renat Fatkhullin #:

Posted documentation on the new features of the built-in matrices and vectors:

• Matrixand Vector Methods
  
• Matrix and vector types
• Initialisation
• Manipulations
• Operations
• Derivations
• Transformations
• Statistics
• Characteristics
• Solutions
• Machine learning

Please go through the descriptions of the functionality.

If you can give us your comments and ideas, it would be very much appreciated.

I have not found initialisation by array values, it would be convenient to get rid of loops.

For example, you need to transfer values from the indicator buffer to a vector or matrix.

I don't seem to have any more questions so far, everything is well described

[Renat Fatkhullin]

Maxim Dmitrievsky #:

I didn't find initialisation by array values, it would be convenient to get rid of loops.

For example, you need to transfer values from the indicator buffer to a vector or matrix

I don't seem to have any more questions so far, everything is well described

The next set of functions is various exchanges between MqlRates, MqlTick, etc., including direct receiving of price data into vectors and matrices.

The article with detailed examples is almost ready.

[Roman]

Renat Fatkhullin #:

The next set of functions is a variety of exchanges between MqlRates, MqlTick, etc., including direct retrieval of price data into vectors and matrices.

The article with detailed examples is almost ready.

Please add methods for vectors
Engle Granger and Dickey-Fuller tests.

And for matrices Johansen test.